This repository contains my portfolio of Machine Learning projects, completed for self-learning and hobby purposes. They are presented in the form of Jupyter Notebooks.

• Machine Learning Projects

  • Binary Classification with the Titanic Dataset: An end to end machine learning project that goes among other things through Data Analysis, Data Preprocessing, building and evaluating Machine Learning models with a deep focus on the Random Forest Classifier, including Hyperparameter tuning.
  • Classification Project: In this Notebook I covered the basics of how supervised binary & multiclass classification works. I got everything I cover in this notebook from the book "Hands-On Machine Learning with Scikit_learn & Tensorflow" from Aurelien Geron. I used the famous mnist dataset, which contains 70,000 images of handwritten digits.
  • Housing Prices with California Housing Dataset: In this notebook I explained how to tackle a machine learning from the beginning to the end, at the example of predicting housing prices. I covered everything you need, to build regression system and a lot of tools & techniques, that are common in the machine learning landscape. I got everything I cover in this notebook from the book "Hands-On Machine Learning with Scikit_learn & Tensorflow" from Aurelien Geron. I used the California Housing Dataset from the statlib repository.
  • machinelearning-blog.com: Setting up my own blog with tutorials and explanations that simplify the concepts of applied Machine Learning.

• Course Certificates

  • Machine Learning Foundations: A Case Study Approach: Certificate of the "Machine Learning Foundations: A Case Study Approach" Coursera course, created from the University of Washington (6 weeks of study). I learned how to predict house prices based on house-level features (Regression), analyze sentiment from user reviews (Classification), retrieve documents of interest (Clustering and Similarity), recommend products, and search for images (Deep Learning).
  • Neural Networks and Deep-Learning: Certificate of the "Neural Networks and Deep-Learning" Coursera course, created from deeplearning.ai and taught by Andrew NG (4 weeks of study). This course covered the foundations of developing deep learning models and the major technology trends driving Deep Learning.
  • Version_Control_with_Git: Certificate of the "Version Control with Git" Coursera course, created from the software company Atlassian (4 weeks of study). Through this course I gained a strong conceptual understanding of Git and learned how to use it in a professional manner.

If you liked what you saw, want to have a chat with me about the portfolio, work opportunities, or collaboration, write an email at niklas.donges@code.berlin.